New Experimental Drivetrain

Hey all,

I had a kinda strange idea for a drivetrain w/ a two 3 cim gearboxes powering the rear wheels and 2 swerve drive modules in the front. My reasoning for this, is that the drivetrain would mimic car steering, giving the advantage of more control at higher speeds, and less friction between wheels and ground; while providing the capability of turning in place. Also it would be easier to much easier to service (i think at least?). Thoughts?

Here is a link to a slideshow that showcases my design…it?usp=sharing

67 did something like that for 2008. They weren’t huge fans of the shopping cart-esque drive.

Honestly if you have the resources to go half swerve, you might as well try out full swerve.

Nikp, the link to your design doesn’t work.

I’d say it is less like a car and more like a fork lift. I had this idea, but I thought about it and quickly ushered it off. Swerve is overly complex, and you are losing the holonomic value that the complexity makes worth while. Also I don’t see why you want your pivot point towards the front/back of your robot. It makes you far less mobile then a drop center pivot point of your standard drive train.

2471 did this last offseason.

525 did something similar last year, with 1 swerve module in the back.

Is anyone else having trouble with the link? I keep on getting error 400

If you want to drive like a car, why not use the same Ackermann mechanisms cars use to drive?

There are compromises in every mechanical Ackermann system, easier to implement no-compromises Ackermann in software than to run an ugly linkage all the way across the robot.

Alternatively, if you don’t want to do a bunch of work programming the independent modules, use a linkage and a steering servo like an RC car to get away from the hard programming.

Or just build an RC car. They’re pretty sweet.

118 did something very much like what you are thinking of in 2009.
Here’s the video,

Another recent example would be 1444 in 2016.